PROJECT SUMMARY Candidate. The training and research plan proposed in this Pathway to Independence Award will propel the candidate to become an independent scientist in a tenure-track position at a research university. This award will support her investigation of novel questions regarding structural connectivity of neural networks that subserve episodic memory across the lifespan. She will be introduced to high-resolution multimodal neuroimaging and will receive training in corresponding advanced MRI and multivariate analysis techniques. The candidate's expertise in neurocognitive aging research will also be strengthened by the proposal's focus on component processes of episodic memory (i.e., pattern separation), the neural networks that support these processes, and how they are affected in both healthy aging and in individuals at increased risk for dementia. Importantly, this award will prepare the candidate to submit a major research proposal (e.g., R01) at an earlier stage in her career that would be possible otherwise. Environment. The University of California, Irvine (UCI) offers a unique array of training and development resources to facilitate the candidate advancing to an independent scientist position. These include a collaborative group of distinguished researchers at the Center for the Neurobiology of Learning and Memory (CNLM) dedicated to understanding neural mechanisms that support memory, an exceptional mentor (Dr. Craig Stark, Director of the CNLM) and co-mentor (Dr. Claudia Kawas, Clinical Core director of the UCI Institute for Memory Impairments and Neurological Disorders) whose pioneering research programs laid the foundation for the current proposal, access to state-of-the-art research and neuroimaging facilities, and a variety of courses and workshops that will accelerate both educational and career development throughout the duration of this award. Research. The central aim of the current proposal is to investigate neural networks of pattern separation, a component process of episodic memory, across the life span using behavioral and high- resolution multimodal neuroimaging techniques. Episodic memory decline is a hallmark feature of healthy aging and age-related cognitive disorders such as amnestic mild cognitive impairment and Alzheimer's disease. Further detailing the neural mechanisms that support component processes of episodic memory may facilitate identification of neural markers associated with cognitive aging, and inform cognitive and neural interventions aimed at promoting successful aging. Episodic memory is a complex mnemonic ability that involves encoding and retrieval of discrete events, including details such as what, where, and when an event occurred. Successful encoding of new information requires that similar events get separated into distinct memory representations. This process, termed pattern separation, is known to rely on medial temporal lobe (MTL) subregions. However, neuroimaging studies have shown that prefrontal cortex (PFC) and striatum are also engaged during episodic memory performance. Whereas these distributed brain regions are frequently studied in isolation, a comprehensive understanding of the neural substrates of episodic memory in general, and pattern separation in particular, will require knowledge of how they interact as interconnected neural networks. In the mentored phase of this award, high-resolution diffusion tensor imaging (DTI) and functional magnetic resonance imaging (fMRI) will be used to identify more accurate models of MTL (Specific Aim 1) and striatal (Specific Aim 2) connectivity across the human lifespan than has been previously acquired in vivo. Expanding on our earlier work with the perforant path, the proposed study will assess the contribution of local tracts connecting MTL subregions (e.g., perforant path, mossy fibers, and schaffer collaterals) and large-scale tracts connecting MTL to PFC (e.g., fornix, cingulum) to pattern separation performance in healthy adults. It will also be the first to examine integrity of tracts connecting striatum to PFC (e.g., caudate-PFC, putamen-motor) in relation to these mnemonic processes. The independent phase of this award will assess interactions and dissociations between MTL and striatal memory systems, which are frequently regarded as being differentially affected by healthy and pathological aging. We will test the hypotheses that degradation of striatal versus MTL tract integrity accounts for pattern separation declines in healthy older versus younger adults (Specific Aim 2) and that degradation of MTL versus striatal tract connectivity accounts for pattern separation declines in oldest- old versus younger-old adults (Specific Aim 3). These data will directly test cortical disconnection theories, which propose that diminished white matter connectivity accounts for cognitive declines associated with aging.